In-house model

Next, we used an in-house library design software (see details in Chapter 15) to enumerate the virtual libraries and then calculated various physical properties. Products were removed from consideration if MW is > 300, number of rotatable bonds > 3, and ClogP > 3. For solubility, two in-house model calculations were applied as filters turbidimetric >10 mg/mL and thermodynamic solubility >100 xM. The resulting cherry-picked library was then reviewed by NMR spectroscopists to remove compounds with possible artifacts, likely to be insoluble, or likely to be false positive. These included some conjugated systems and compounds with likelihood of indistinct NMR spectra. 

Novartis uses the In Silico Profiling web tool. Available properties include the octanol water partition coefficient log P, molar refractivity, flexibility index, hydrogen bonding characteristics and molecular polar surface area. Various drug properties, such as intestinal absorption, BBB permeability or Plasma Protein Binding (PPB) are calculated based on in-house models. 

The largest commercially available datasets are the Physical Properties (PHYSPROP) and AQUASOL databases ca. 6000 compounds in each database). The AQUASOL database has been published as a book. Furthermore, two relatively large sets of aqueous solubility data models were used in many other studies.Data from the AQUASOL database had an interlaboratory variation of about a = 0.49 log-units (as estimated for A=1031 molecules).Moreover, large inter-laboratory errors mask the influence of temperature, and differences as large as AT = 30 °C do not increase this error. In-house models developed at pharmaceutical companies could be based on similar or even larger numbers of measurements. For example, about 5000 molecules were used to develop a model at Bayer Healthcare AG. " ... 

Second, the integration of mathematical models for unit operations into commercial process simulators has been carried forward with unit operation and thermodjmamic models from research at universities as well as in-house models provided by end users. Background knowledge and experiences gained in the various integration efforts in the IMPROVF project have helped to effectively develop these solutions. From a technical perspective, solutions are based on the independent CAPE-OPEN standard rather than on proprietary interfaces of the various simulators in which the models can be employed. 

Distribution) most economical loads connection for in-house model, Windows NT server... 

Other models exist where the keratinocyte cultures grown at air-liquid interface are cultured in different substrates the SkinEthic model (Rosdy and Clauss, 1990), the Cosmital in-house model (Wella, Switzerland) (Faller et ai, 2002), and the Apligraf model (Organogenesis Inc., Novartis Pharmaceuticals Corporation, Canton, MA USA) (Medina et al, 2000). The SkinEthic is not a validated method and the other models need further development. 

The use of color graphics is also an effective means for displaying chemical stmctures. This method is far better than typesetting the three-dimensional architecture of complex multimolecule assembly (112). For developing in-house CAD software programs, the three-dimensional, sohd-modeling capabiUties of SdverScreen can also be utilized either in monochrome or color for constmction of such stmctures (113). 

If the hypothesis or model does not seem to be a good predictor of what is happening in the building, you probably need to collect more information about the occupants, HVAC system, pollutant pathways, or contaminant sources. Under some circumstances, detailed or sophisticated measurements of pollutant concentrations or ventilation quantities may be required. Outside assistance may be needed if repeated efforts fail to produce a successful hypothesis or if the information required calls for instruments and procedures that are not available in-house. Analysis of the information collected during the LAQ investigation could produce any of the following results ... 

Hunt and Kulmala have solved the full turbulent fluid flow for the Aaberg system using the k-e turbulent model or a variation of it as described in Chapter 13— the solution algorithm SIMPLE, the QUICK scheme, etc. Both commercial software and in-house-developed codes have been employed, and all the investigators have produced very similar findings. 

Our whole package Is In-house specified and written, and while we are aware that complete packages to data-log are now available, we feel confident In showing this as an example of a high technology state-of-the-art system. Examples of data extracted for heat balance, and comparison to simple models can be demonstrated. 

Muscarinic M3 Receptor. A pharmacophore model was derived from known M3 receptor antagonists, using the program DISCO, and 3D searching was performed by Unity 3D in the Astra Charnwood in-house compound repository and the databases of several commercial suppliers. The 172 compounds that fitted the pharmacophore were screened for their M3-antagonistic potency. Several compounds with micromolar and even submicromolar activities resulted, for example, compound 13 (A50 M3 antagonism 0.2pM pA2 = 6.67 Fig. 16.2) [85],... 

A CRO may also allow for the in-house introduction of specialized lipophilic scales by transferring routine measurements. While the octanol-water scale is widely applied, it may be advantageous to utilize alternative scales for specific QSAR models. Solvent systems such as alkane or chloroform and biomimetic stationary phases on HPLC columns have both been advocated. Seydel [65] recently reviewed the suitabihty of various systems to describe partitioning into membranes. Through several examples, he concludes that drug-membrane interaction as it relates to transport, distribution and efficacy cannot be well characterized by partition coefficients in bulk solvents alone, including octanol. However, octanol-water partition coefficients will persist in valuable databases and decades of QSAR studies. 

Considering the failure of available log D methods to predict in-house data and taking into account that such data are usually generated just for a few fixed pH values, a number of companies started to elaborate in-house methods for log D prediction at fixed pH. Up to date several companies have reported development of such methods. For example, Cerep has developed methods to predict log D at pH 7.4 and 6.5 included in their Bio Print package [107], but details of their method are not pubhshed. HQSAR Tripos descriptors were used by Bayer to develop log D models at pH 2.3 and 7.5 using 70000 (qi =0.76, STD =0.60) and 7000 (qi =0.83, STD =0.67) compounds, respectively [108] however, again, no details of the approach were provided. 

In order to be appUed the models should be predictive. Unfortunately, the models frequently fail and demonstrate significantly lower prediction ability compared to the estimated one, when they are applied to new unseen data [100-103, 106]. One of the main reasons for such failures can be the lack of available experimental data and difficulties in calculating log D, as discussed in Section 16.4.2. Another problem of low prediction ability of log D models can be attributed to different chemical diversity of molecules in the in-house databases compared to the training sets used to develop the programs. 

A low accuracy of models for prediction of log D at any pH would not encourage the use of these models for practical applications in industry. Thus, it is likely that the methods for log D prediction at fixed pH that are developed in house by pharmaceutical companies will dominate in industry. However, log D measurements... 

Muresan, S., Sadowski,). In-house likeness comparison of large compound collections using artificial neural networks. f Chem. Inf Model. 2005, 45, 888-893. 

---